Cancer Research Centre Aims to Push the Boundaries of Cancer Research with AI Support

The potential of artificial intelligence to enhance both the speed and accuracy of identifying imaging biomarkers is well recognised. As part of a research initiative, The Royal Marsden has developed and evaluated AI algorithms designed to improve the accuracy of cancer detection. This specifically involves cancers such as sarcomas, lung, breast, brain, and prostate cancer. The result will be faster response times and more accurate diagnoses.

Once the service is live, research teams at The Royal Marsden will be able to evaluate a range of AI models for different types of cancer. These studies will provide crucial insights into how AI can support clinical decision-making and shape future approaches to diagnosis and treatment. Through CARPL.ai’s centralised interface and integrated monitoring tools, researchers will also be able to track model performance over time, enabling faster feedback loops between development and deployment – a vital step towards clinical implementation.

Collection of AI Models for Radiology

The new service runs on a clinical MLOps imaging platform, built and operated by NTT DATA. The AI computing solution leverages high-performance Dell servers and the latest GPU processing power – orchestrated by the CARPL.ai platform, which assists with testing and managing AI algorithms and includes a large repository of AI models for radiology.

“The new service is a prime example of the innovative power of AI, and specifically of its ethical and safe application in healthcare,” says Oliver Köth, CTO of NTT DATA. “In addition to the MLOps platform for image analysis, we are also supporting researchers at The Royal Marsden in testing new AI tools in real hospital settings and in deriving insights from various algorithms.”